Following the Herds: The Buffalo and the Caribou

Status: Researched draft — cultural review pending

How to read this entry: Documented fact = archaeological record, government or museum publication, or published scholarship (Sources). Tradition — the people's own account = oral history or a people's own published teaching, labelled as such. Debated / Claim = disputed point, shown as a sourced range with the disagreement printed. No ceremonial or restricted content is described. This entry is a history in time — from earliest evidence through height to disruption and continuance — not a catalogue of techniques. A practice named for one people is not transferred to another.

The Buffalo: Paleoindian Beginnings

• Clovis and Folsom. Documented fact: The Clovis tradition, about 13,200–12,900 years before present, is associated on the Plains with hunting large Pleistocene game, including the extinct Bison antiquus (University of Colorado Museum of Natural History, 2023). The later Folsom tradition based its Plains subsistence substantially on extinct bison, as at the Folsom site, New Mexico, and Lindenmeier, Colorado — a shift toward organised, repeated taking of herds (EBSCO Research Starters, n.d., summarising the excavations).
• Olsen-Chubbuck. Documented fact: At Olsen-Chubbuck, Colorado, Joe Ben Wheat (excavations 1958, 1960) documented a Cody-complex kill dated about 8000–6500 BCE: nearly 200 bison driven into a dry arroyo, killed and butchered (Wheat, 1972, The Olsen-Chubbuck Site, Society for American Archaeology Memoir 26; University of Colorado, 2023). Wheat estimated about three-quarters were butchered, yielding roughly 56,000–60,000 pounds of meat, enough on his reconstruction for about 100–150 people for a month, with some meat dried. Debated / Claim: head-counts in summaries vary (c. 157 to c. 190–200) and the people-fed figure is a bone-bed reconstruction, not an observed count.
• The pattern set early. Documented fact: Long before horses or firearms, Plains hunters were reading herd behaviour, using topography — arroyos, and later cliffs and corrals — as the killing instrument, and processing at a scale requiring many hands (Wheat, 1972; Frison, 1978, Prehistoric Hunters of the High Plains). Jump, pound and surround all develop those three elements.

The Jump Tradition and Head-Smashed-In

• Length of use. Documented fact: Head-Smashed-In Buffalo Jump, in Blackfoot territory in present-day southwestern Alberta, was in use from about 5,800 years before present to about 1850 CE — nearly 6,000 years (Parks Canada, n.d., statement of significance; UNESCO, 1981, World Heritage inscription). Parks Canada calls it one of the oldest and largest jumps on the northern Plains, with bison bone deposits over 10 metres deep below the cliff.
• Tradition — the people's own account: The Blackfoot name is Estipah-skikikini-kots. The English name derives from a Blackfoot account of a young man who stood below the cliff to watch, was killed by falling bison, and was found with his head smashed in — recorded here as the account as Parks Canada and the site programme tell it, not as an archaeological finding (Parks Canada, n.d.).
• Gathering basin and drive lanes. Documented fact: The jump was a system, not a cliff alone: a broad gathering basin where herds were slowly collected, and drive lanes of stone cairns, extending in places for kilometres, converging in a V toward a cliff edge concealed until animals were upon it (Parks Canada, n.d.; Brink, 2008, Imagining Head-Smashed-In). Parks Canada calls this basin-and-lane complex among the largest and most complex known, and essentially undisturbed.
• The buffalo runner. Documented fact: Site and ethnographic accounts describe a specialised front role: a young, fast runner in a calf skin imitated a distressed calf to draw lead animals into the lane, while hunters in wolf skins appeared at the flanks and rear to turn unease into a stampede; people hidden behind cairns then rose, shouting and waving hides, to hold the herd in the lanes (Brink, 2008; Grinnell, 1892, Blackfoot Lodge Tales). The runner escaped at the edge to a safe ledge. The role rested on detailed knowledge of bison senses — poor forward vision, strong herd-following, and reaction to motion and scent (Brink, 2005; 2008).
• The processing camp. Documented fact: Below the cliff lay the kill area and a processing camp — tipi rings, tools and butchering implements are part of the designated site, with artefacts from about 4800 BCE onward — where meat was cut and dried, hides dressed, and bone rendered for grease (Parks Canada, n.d.). A jump thus supplied both immediate distribution and stored food for winter.
• A wide tradition. Documented fact: Old Women's Buffalo Jump (Aakíípisskan), near Cayley, Alberta, was used for about 2,000 years and was the first major jump excavation on the Canadian Prairies, in 1958–59 (Alberta historic-places documentation, 2016). Jumps were a northern-Plains system over millennia, not a single invention.

Pounds, Surrounds and Running Hunts

• The pound — piskun. Documented fact: Where no cliff served, hunters built a pound: a strong corral of logs, brush and hides at a coulee foot or lane end, into which bison were funnelled and killed at close quarters. The Blackfoot word piskun / pis'kun is glossed "deep blood kettle" (Dictionary of Canadianisms on Historical Principles, 3rd ed., "piskun"; Grinnell, 1892). Grinnell, from Blackfoot informants, describes the pound built out from a bluff foot, with two diverging lines of rock piles and brush forming the V above it, and the 1952 source cited in the dictionary notes the pound as a communal enterprise whose kill was distributed equally (Grinnell, 1892; DCHP-3).
• Who used pounds. Documented fact: The historic summary lists among pound- and jump-users Blackfeet, Plains Cree, Gros Ventre (Aaniiih), Sarcee (Tsuut'ina) and some Dakota bands, with Cheyenne and Arapaho building pounds on level ground (Grinnell, 1892). The technology was shared; each people built and worked it in its own way.
• Surrounds and seasons. Documented fact: In a surround, a widening circle of people closed around a herd, frightening animals into milling until they tired and could be killed; breaking the circle early scattered the herd and ruined the hunt for all (Grinnell, 1892). Alberta's communal-hunting survey also lists driving herds with fire, onto frozen lakes, and into deep snow as documented methods (Alberta Historic Places, 2016).

Horses, the Robe Trade and the Pemmican Economy

• The horse. Documented fact: Horses spreading north from Spanish sources made the running hunt — pursuing a herd at the gallop — a major method by the 18th century, alongside, not instantly replacing, jumps and pounds, which continued on the northern Plains into the 19th century (Isenberg, 2000, The Destruction of the Bison; Parks Canada, n.d., use at Head-Smashed-In to c. 1850).
• Pemmican production. Documented fact: Pemmican (Cree pimîhkân, from pimî, fat/grease) was made in the processing camps, principally by women: bison meat cut thin, dried on racks two to three days, pounded on a hide to powder, mixed with hot rendered fat and marrow fat and, when available, dried berries, then poured into bison-hide bags of about 90–100 pounds, sewn shut and cooled (Colpitts, 2015, Pemmican Empire; Hargrave, 1871, Red River, on the Métis bag). This is a published historical summary only, not a recipe.
• Pemmican and the fur economy. Documented fact: Because a bag kept for months or years and concentrated several times its weight in fresh meat, pemmican fuelled the northern fur trade: Hudson's Bay Company and North West Company brigades could not travel the Saskatchewan and northern routes on imported food alone, and Assiniboine and Cree producers were supplying dried meat and grease to posts from the late 1600s–1700s (Colpitts, 2015; Ray, 1974, Indians in the Fur Trade). The seasonal logic followed the herd: summer and autumn aggregation hunts produced the surplus; winter produced prime robes.
• The robe trade, 1830s–1860s. Documented fact: When beaver demand fell and steamboats reached the upper Missouri in 1832, heavy winter robes — hides taken November to February, dressed with hair on — became a bulk export for the first time (Isenberg, 2000; Lewis, 1942). Lewis records hides traded at Fort Piegan rising from about 2,000 in 1834 to about 20,000 by 1841 in Blackfoot country, with documented consequences including intensified hunting and changes in household labour (Lewis, 1942, cited in Lycett & Keyser, 2021). Debated / Claim: robe totals vary by company, post and year and survive incompletely; figures here are single-post/company figures, not continental totals.

The Métis Organised Hunts and the Laws of the Hunt

• From family hunts to brigades. Documented fact: Before about 1816 Métis hunting from Red River was largely small-scale and familial; as demand for pemmican, meat and hides grew, the hunts became large, twice-yearly brigades governed for their duration by elected officers (Métis Museum / Gabriel Dumont Institute, n.d., "Bison Hunting"; Hargrave, 1871).
• Hunt government. Documented fact: The assembly elected a chief of the hunt and ten captains, each choosing ten soldiers to patrol in turn and enforce the rules (Métis Museum, n.d.). Métis historians treat this as a root of later self-government: in 1873 the Southbranch settlements organised local government under Gabriel Dumont on the hunt-law model (Métis Museum, n.d.; Indigenous Peoples Atlas of Canada, n.d.).
• The Laws of the Hunt, recorded 1840. Documented fact: The code — the Law of the Prairie — provided: no buffalo to be run on the Sabbath; no party to fork off, lag or go before without permission; no person to run buffalo before the general order; captains to patrol and guard in turn; a first offence punished by cutting up the offender's saddle and bridle, a second his coat, a third by flogging; and a thief, "even to the value of a sinew," to be named aloud three times as "thief" by the camp crier (Métis Museum, n.d.). Its stated purposes were brigade safety and preservation of the herd against a premature run that scattered the common stock (Métis Museum, n.d.).
• Height and conflict. Documented fact: The great hunts of the 1840s–70s brought Métis brigades into contest with Dakota (Yanktonai) parties, culminating in the Battle of Grand Coteau, 13–14 July 1851, after which the two nations made peace (Indigenous Peoples Atlas of Canada, n.d.). The hunt was at once a food economy, a trade enterprise and a national undertaking — which is why its collapse struck all three at once.

Destruction, 1870s–1880s

• The hide hunt. Documented fact: After the American Civil War, an industrial hide market in the eastern United States and Europe, long-range rifles and railroads produced a brief, intensive commercial slaughter, mostly 1871–1883; the southern herd was largely destroyed in about four years, the heaviest phase in 1872–74, hunters typically taking the hide and leaving the carcass (Isenberg, 2000; EBSCO Research Starters, n.d., "Bison slaughter impact").
• United States policy. Documented fact: Senior U.S. Army officers and members of Congress stated on the record that destroying the buffalo would compel Plains nations onto reservations by removing their food base, and the Army aided hide hunters, including, in the EBSCO summary, supplying ammunition; federal protection bills were defeated or vetoed in the 1870s while the slaughter proceeded (Isenberg, 2000; Hornaday, 1889). Debated / Claim: General Philip Sheridan's endorsement of the hunters is attested in contemporary reporting of his 1875 remarks opposing a Texas protection bill, but its exact wording varies between sources and is not asserted here.
• How many, and how few. Debated / Claim — ranges, disagreement printed: Pre-destruction estimates in the literature range from about 30 million to 60 million, with older popular accounts higher; no census existed (Isenberg, 2000; Hornaday, 1889). The endpoint was counted: William T. Hornaday's Smithsonian census as of 1 January 1889 totalled 1,091 bison in all North America, wild and captive — including about 550 wood bison in Athabasca, about 200 protected in Yellowstone, and only 85 free-ranging in the United States outside Yellowstone (Hornaday, 1889, The Extermination of the American Bison). In Canada the plains herds were effectively gone by about 1880; a following bone trade included one buyer's estimate that he alone bought skeletons of about six million bison, 1884–91, for fertiliser — flagged as a single buyer's estimate (EBSCO, n.d.).
• Consequence. Documented fact: For Blackfoot, Lakota, Cheyenne, Arapaho, Cree, Assiniboine, Métis and other Plains peoples, losing the herd within a decade removed food, clothing, shelter and trade income at once, and was a direct instrument — in Canada as in the United States — in pressing hungry communities into treaty, reserve and reservation settlements in the late 1870s–1880s (Isenberg, 2000; Daschuk, 2013, Clearing the Plains).

Restoration by Nations

• Survival. Documented fact: The species survived in the Yellowstone herd, the Athabasca wood bison, a few private ranch herds (Goodnight in Texas; Pablo–Allard in Montana) and zoo herds from Hornaday-era rescues; some of the earliest reservation herds date to the 1930s, including Pine Ridge and Crow (Hornaday, 1889; Isenberg, 2000).
• InterTribal Buffalo Council. Documented fact: Founded in 1992 by a small group of tribes and federally chartered in 2009, the ITBC states its mission as "restoring buffalo to Indian Country, to preserve our historical, cultural, traditional and spiritual relationship for future generations" (ITBC, n.d.; Carlson, 2026, testimony to the U.S. House Appropriations Subcommittee). Its figures are time-stamped: 57 member tribes in 2009; 69 tribes in 19 states with over 20,000 buffalo at its 30-year mark; 89–90 tribes in 22 states managing over 25,000 buffalo in 2025–26 reporting; and over 1,500 buffalo transferred to more than 20 nations in 2025 alone, with National Park Service, Fish and Wildlife Service and Nature Conservancy partners (ITBC, 2025; Carlson, 2026). In Canada, plains bison have also returned to First Nations lands and, in a Parks Canada reintroduction, to Banff National Park in 2017 — each herd is a separate nation's undertaking, to be named on its own page at publication. Flagged for publication check.

The Caribou: Built on Migration

Caribou systems were built not on a resident herd but on migration: twice-yearly movements between calving and wintering grounds, crossing the same rivers, lakes and passes at predictable seasons. The historical systems therefore centred on interception — knowing where the herd must pass, and being there.

Innu and the George River Herd

• Antiquity. Documented fact: Innu and their ancestors have hunted caribou (atiku) in Nitassinan, the Labrador–Quebec peninsula, for millennia — some 8,000 years in the published synthesis used for the Innu record (Samson & Pretty, 2006; Tipatshimuna, Innu Nation / Canadian Museum of History, n.d.). Caribou was the central interior game, alongside bear, beaver, fish and waterfowl.
• Interception at crossings. Tradition — the people's own account: The oldest Innu elders remembered caribou speared from canoes as they swam the Mushuau-shipu (George River), and multi-family shaputuan camps placed for the autumn hunt (Tipatshimuna, n.d., elders' accounts). Parks Canada's Kameshtashtan (Mistastin Lake) designation records the same pattern archaeologically: camps near caribou drive locations and water-crossing ambush places, caches and hearths, on a route the George River herd passes twice yearly (Parks Canada, n.d.).
• Innu law respecting caribou. Documented fact: The ethnohistorical record documents extensive Innu laws on respect for caribou — treatment of meat, fat, marrow, bones and antlers — and the Caribou Master, Kanipinikassikueu, who could withdraw caribou from hunters who violated them (Speck, 1935, Naskapi; Tipatshimuna, n.d.). This entry records that the law exists and governed the hunt; it does not reconstruct ceremonial practice from outside sources.
• Disruption. Documented fact: In the mid-20th century Innu families were pressed into settled villages — including the Mushuau Innu relocation to Davis Inlet in the 1950s, when caribou were scarce — ending the full-time nomadic round for most families within a generation, though families continued returning to the country to hunt (Samson & Pretty, 2006; Tipatshimuna, n.d.).

Dene Drive Fences and Corrals

• The system. Documented fact: In the Mackenzie Mountains and Sahtu, Shúhtagot'ine (Mountain Dene) and Sahtu Dene hunters built long wooden fences following ridges and descending into valleys, channelling migrating caribou toward a corral or kill zone — the forest analogue of the Plains drive lane (Beckhusen et al., 2022; Andrews, Prince of Wales Northern Heritage Centre, 2016).
• Moose Horn Pass, dated. Documented fact: The Moose Horn Pass Caribou Fence (KjRx-1), in Shúhtagot'ine homeland, comprises three fences; the main fence is nearly 800 metres long and ends in a corral in the valley. Tree-ring dating of 62 fence timbers, against a chronology from 972 to 2016 CE, gave terminal dates spanning 1314–1876 CE, clustering about 1420–1480 and 1580–1750 — a complex built and rebuilt over centuries (Beckhusen et al., 2022, Forests).
• Scale. Documented fact: About 47 caribou fences are recorded in the Northwest Territories, including a Sahtu fence about a kilometre long, about a century old when documented by drone in 2016, whose builders are thought to have supplied meat to Hudson's Bay Company posts in the late 1800s; a further half-dozen are recorded in Vuntut National Park, Yukon, in Gwich'in country (Andrews, 2016). Debated / Claim: the assessment that the strategy was used "for thousands of years" is an archaeologist's inference; dated wood spans only the last several centuries, because wood decays.
• Water crossings. Documented fact: Across the Subarctic, fences on land paired with crossings on water, where swimming caribou were hunted from canoes or shore blinds at known places — documented for Dene peoples as for Innu and Inuit, each people's places named in its own land-use record (Speck, 1935; Stewart et al., 2000).

Inuit Systems: Crossing Places and Inukhuit

• Terminology. Documented fact: The stone figures of the drives are inukhuit (singular inuksuk), "acting in the capacity of a human" in Norman Hallendy's formulation — rows of cairns, sometimes topped with sod, that caribou at distance perceive as a line of people (Hallendy, 1994, cited in Stewart et al., 2000; Brink, 2005). Popular use of "inukshuk" for any stacked stones flattens this working infrastructure.
• How a drive worked. Documented fact: On southern Victoria Island, Nunavut, archaeologists documented the Eggington site — a single-line drive through a saddle between hills to shooting pits — and the POD site, a V-shaped funnel of converging cairn lines and stone walls ending in opposing shooting pits (Brink, 2005, Arctic Anthropology). Drives manipulated caribou reliance on sight, scent and the animal ahead, with terrain, wind and human beaters completing the funnel, sometimes toward a water crossing (Brink, 2005; Arima, 1984, cited there).
• Crossing places as landscapes. Documented fact: On the Kazan River, in Caribou Inuit country, survey recorded about 1,500 features around a series of crossings — blinds above exit points, caches, dwellings concentrated away from the entry shore; only about 14% stood on the north (entry) side (Stewart et al., 2000, Journal of Anthropological Archaeology). The authors connect that pattern to practice voiced by Elder Barnabus Piryuaq: keeping the entry shore clean of caches, skinning places and anything that might make caribou suspicious (Piryuaq, in Mannik, 1993, cited in Stewart et al., 2000). Stone drive lines lead to the water at two north-shore sites.
• Living use. Documented fact: The Netsilik film At the Caribou Crossing Place (National Film Board, 1967, filmed 1963–65 near Kuugaruk) shows Netsilingmiut men building a row of inuksuit to deflect caribou into water, where they were harvested and floated ashore — the system in use within the documented period.

Herd Collapses in the Modern Record

• George River — the counted collapse. Documented fact: The George River herd — hunted by Innu, Nunavik Inuit and Labrador Inuit — grew from about 15,000 in the 1950s to an estimated approximately 800,000 in the early 1990s, then declined by about 99%: about 14,200 (2014), 8,938 (July 2016), about 5,500 (2018) and 8,100 (2020) in the public survey record (Government of Newfoundland and Labrador releases, 2017–2020; Bergerud et al., 2008). The province imposed a total hunting ban in 2013, still in place through those releases.
• Causes — as the record states them. Debated / Claim: The provincial backgrounder attributes the decline's start to habitat and food deterioration after the very high early-1990s population, with predation, hunting and possibly climate change growing in significance afterward (Government of Newfoundland and Labrador, 2017). Inuit-led research records that exact causes are not fully understood, and elders hold — Tradition — the people's own account — that caribou cycle over decades and have recovered from lows before, a position standing beside the province's extirpation-risk framing (Borish et al., 2021, Global Environmental Change; HERD project, 2016–22). Neither position is settled here.
• Wider pattern. Documented fact: George River is the best-counted case of a wider circumpolar pattern: other migratory herds hunted by Dene, Gwich'in, Inuit and other nations, Bathurst among them, also fell sharply from late-20th-century highs, with co-management restrictions following (Arctic Council / CAFF assessments; territorial wildlife reports). Herd-by-herd figures are not asserted here without their own dated sources.

Cross-reference — the wolf of the herds' country

The herds this page follows were hunted, in their own country, by the wolf as by the peoples — and the wolf of the Southwest and Mexico was destroyed by a government eradication campaign (bounties from the 1890s; the federal PARC programme from 1915; over 900 Mexican wolves counted killed by government trappers and cooperators in Arizona and New Mexico in 1915–1925 alone) and returned from seven founders in a binational recovery in which the White Mountain Apache Tribe stands as a governing partner. That record stands as this doorway's one species-recovery line, at the Extinction Register's Mexican gray wolf — a recovery entry, which cross-refers back to this page. Record: THE_HISTORIAN_COMPLETE_2026-10-30.md §2, P-7.

Sources

• Alberta Historic Places (2016). "Blood Kettles and Buffalo Jumps: Communal Hunting on the Plains of Alberta." Government of Alberta.
• Andrews, Thomas, Prince of Wales Northern Heritage Centre (2016). Sahtu caribou fence documentation, reported CBC News / Archaeology Magazine, Dec. 2016.
• Beckhusen, G. et al. (2022). "Dendroarchaeology in the Mackenzie Mountains: The Moose Horn Pass Caribou Fence (KjRx-1)." Forests.
• Bergerud, A. T. et al. (2008). George River herd history, as cited in Government of Newfoundland and Labrador releases.
• Borish, D. et al. (2021). "'Caribou was the reason, and everything else happened after'." Global Environmental Change.
• Brink, J. W. (2005). "Inukshuk: Caribou Drive Lanes on Southern Victoria Island." Arctic Anthropology; (2008) Imagining Head-Smashed-In. Athabasca University Press.
• Colpitts, G. (2015). Pemmican Empire. Cambridge University Press.
• Daschuk, J. (2013). Clearing the Plains. University of Regina Press.
• Dictionary of Canadianisms on Historical Principles, 3rd ed. (n.d.). "piskun." UBC.
• Frison, G. C. (1978). Prehistoric Hunters of the High Plains. Academic Press.
• Government of Newfoundland and Labrador (2017–2020). George River herd releases and backgrounder, 2 Feb. 2017, and survey releases.
• Grinnell, G. B. (1892). Blackfoot Lodge Tales. Scribner's.
• Hargrave, J. J. (1871). Red River. John Lovell.
• Hornaday, W. T. (1889). The Extermination of the American Bison. Smithsonian Institution.
• Indigenous Peoples Atlas of Canada (n.d.). "Early Nationalism." indigenouspeoplesatlasofcanada.ca.
• InterTribal Buffalo Council (n.d.) mission; (2025) annual transfer release; Carlson, E. (2026), testimony, U.S. House Appropriations Subcommittee, 18 Mar. 2026.
• Isenberg, A. C. (2000). The Destruction of the Bison. Cambridge University Press.
• Lewis, O. (1942). The Effects of White Contact upon Blackfoot Culture. AES Monograph 6.
• Métis Museum / Gabriel Dumont Institute (n.d.). "Bison Hunting." metismuseum.ca.
• National Film Board of Canada (1967). At the Caribou Crossing Place.
• Parks Canada (n.d.). Head-Smashed-In statement of significance; Kameshtashtan designation.
• Ray, A. J. (1974). Indians in the Fur Trade. University of Toronto Press.
• Samson, C. & Pretty, J. (2006). Innu hunting lifestyles and diets synthesis. Food Policy.
• Speck, F. G. (1935). Naskapi. University of Oklahoma Press.
• Stewart, A. M.; Keith, D.; Scottie, J. (2000). "Caribou Crossings and Cultural Meanings." Journal of Anthropological Archaeology.
• Tipatshimuna — Innu Nation / Canadian Museum of History (n.d.). Elders' accounts. tipatshimuna.ca.
• UNESCO (1981). Head-Smashed-In Buffalo Jump, World Heritage inscription.
• University of Colorado Museum of Natural History (2023). "Bison, the Sustainer of Early Life."
• Wheat, J. B. (1972). The Olsen-Chubbuck Site. SAA Memoir 26.

Weakest-sourced / flagged points in this entry (flagged): (1) Pre-destruction bison population (30–60 million) — reconstructions, no census; disagreement printed. (2) Olsen-Chubbuck counts and people-fed figures — Wheat's reconstructions; summaries differ. (3) Sheridan's wording — sentiment attested, verbatim varies, not asserted. (4) Six-million-skeletons bone figure — single buyer's estimate, secondarily summarised. (5) Robe totals — single-post/company figures, not continental totals. (6) Dene fence antiquity "thousands of years" — inference; dated wood is 1314–1876 CE. (7) ITBC totals — dated figures printed because the council is growing. (8) Canadian First Nations herd restorations beyond Banff (2017) — general terms only, pending herd-by-herd dated sources. (9) George River decline causes — provincial, scientific and elder accounts differ; all printed, none settled.